Predicting Drug-Drug Interactions Using Knowledge Graphs
Preprint
Abstract
In the last decades, people have been consuming and combining more drugs than before, increasing the number of Drug-Drug Interactions (DDIs). To predict unknown DDIs, recently, studies started incorporating Knowledge Graphs (KGs) since they are able to capture the relationships among entities providing better drug representations than using a single drug property. In this paper, we propose the medicX end-to-end framework that integrates several drug features from public drug repositories into a KG and embeds the nodes in the graph using various translation, factorisation and Neural Network (NN) based KG Embedding (KGE) methods. Ultimately, we use a Machine Learning (ML) algorithm that predicts unknown DDIs. Among the different translation and factorisation-based KGE models, we found that the best performing combination was the ComplEx embedding method with a Long Short-Term Memory (LSTM) network, which obtained an F1-score of 95.19% on a dataset based on the DDIs found in DrugBank version 5.1.8. This score is 5.61% better than the state-of-the-art model DeepDDI. Additionally, we also developed a graph auto-encoder model that uses a Graph Neural Network (GNN), which achieved an F1-score of 91.94%. Consequently, GNNs have demonstrated a stronger ability to mine the underlying semantics of the KG than the ComplEx model, and thus using higher dimension embeddings within the GNN can lead to state-of-the-art performance.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- PharmKG: a dedicated knowledge graph benchmark for bomedical data miningBriefings in Bioinformatics, 2020
- Drug-Drug Interaction Predicting by Neural Network Using Integrated SimilarityScientific Reports, 2019
- Modeling polypharmacy side effects with graph convolutional networksBioinformatics, 2018
- Deep learning improves prediction of drug–drug and drug–food interactionsProceedings of the National Academy of Sciences, 2018
- Large-scale structural and textual similarity-based mining of knowledge graph to predict drug–drug interactionsJournal of Web Semantics, 2017
- Toward a complete dataset of drug–drug interaction information from publicly available sourcesJournal of Biomedical Informatics, 2015
- Pharmacogenomics Knowledge for Personalized MedicineClinical Pharmacology & Therapeutics, 2012
- INDI: a computational framework for inferring drug interactions and their associated recommendationsMolecular Systems Biology, 2012
- Bio2RDF: Towards a mashup to build bioinformatics knowledge systemsJournal of Biomedical Informatics, 2008
- KEGG: Kyoto Encyclopedia of Genes and GenomesNucleic Acids Research, 2000